Deep learning on compressed sensing measurements in pneumonia detection

Abstract

Pneumonia is one of the very common life-threatening diseases and needs proper diagnosis at an early stage to be cured expeditiously. Medical practitioners use chest X-ray as the best imaging modality to identify pneumonia. Due to the limited facilities available at the remote places and the need of maintaining the social distancing imposed by the recent outbreak of coronavirus disease, one may not have ease of access to a professional radiologist. This article proposes a deep learning (DL) framework that detects pneumonia from X-ray images to assist the medical practitioners located at distant places. The X-ray images are captured as compressed sensing (CS) measurements i.e. very few numbers of samples are observed in order to obtain an energy efficient and bandwidth preserving system to be utilized for far-end pneumonia detection purpose. Extensive simulation results show that the proposed approach enables the detection of pneumonia with 96.48% accuracy when only 30% samples are transmitted.

Pneumonia Detection Technique Empowered with Transfer Learning Approach

“…Different methodologies have already been applied to the problem of chest X-ray pneumonia. Researchers have utilized various algorithms and CNN models to classify chest X-rays (Islam et al, 2022). The proposed (TDPLA) model achieved higher accuracy than other studies.…”

Comparison of Common Algorithms for Single-Pixel Imaging via Compressed Sensing

“…There are three types of measurement matrices for compressive sensing SPI [ 46 ]: random measurement matrices [ 47 , 61 , 62 , 63 ], partial orthogonal measurement matrices [ 47 , 64 , 65 , 66 ], and semi-deterministic random measurement matrices [ 64 , 67 , 68 , 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 ]. In addition, there are some optimized measurement matrices proposed for improving the RIP characteristics of the measurement matrix [ 77 , 78 , 79 , 80 , 81 , 82 , 83 , 84 , 85 ] or machine learning [ 86 , 87 , 88 , 89 ].…”

“…In addition, methods for matrix optimization are also proposed, such as an optimal projection matrix that optimizes the measurement matrix by using the sparse dictionary of signals [ 84 ] and an optimal measurement matrix based on effective projection [ 85 ]. Researchers are also trying to use machine learning to train and optimize the sampling framework of CS [ 86 , 87 , 88 , 89 ].…”

Explanatory classification of CXR images into COVID-19, Pneumonia and Tuberculosis using deep learning and XAI

“…COVID-19 has been a severe pandemic and the exponential growth of COVID-19 cases has made tremendous pressure on the healthcare system around the globe. Similarly, Pneumonia and TB are also life-threatening diseases [3] , [4] . As a result, accurate and timely detection of such diseases is critical for appropriate treatment and saving lives [5] , [6] .…”